AN ABSTRACT OF THE THESIS OF
Methea Sapp for the degree of Masters of Science in Zoology presented on
Defense Date. August 24th1, 2004
Title: Dinosaur Lung Structure and Ventilalion of the Abdominal Air Sacs in Birds.
Redacted for Privacy
Abstract approved:
The purpose of this study was to identify any ostelogical features which might prevent
paradoxical movement (=lateral collapse) of the abdominal air sacs in birds during inhalation.
A combination of 26 fresh and frozen adult bird carcasses representing 0 avian orders were
procured from local sources. Dissections of each specimen first entailed the inflation of the
respiratory system via a tracheal tube. The trachea was then sealed off to prevent deflation
of the air sacs. Next, the abdominal and thoracic cavities were carefully dissected to expose
the inflated air sacs. Using a set of Whitworth electronic digital calipers, a series of two
measurements were taken on each carcass: I. from the posterior sternal tip to the posterior
end of the abdominal air sacs, and 2. from the posterior sternal tip to the tip of the pubic bone.
Results indicated a strong correlation between the post-sternal length and position of the
abdominal air sacs and the post-sternal extension of the pubic bones, such that lateral support
of the abdominal air sacs was provided by the pubic bones. These data suggest that during
inhalation the pubic bones probably serve to prevent the lateral collapse of these air sacs. The
pubic bones of theropod dinosaurs, and even
Archaeopteryx,
were vertically orientated and
distally fused. Therefore, it is unlikely that these taxa ventilated a set of abdominal air sacs
and where unlikely to have possessed an avian lung-air sac system.
©Copyright by Methea Sapp
August 24th, 2004
All Rights Reserved
Dinosaur Lung Structure and Ventilation of the Abdominal Air Sacs in Birds
by
Methea Sapp
A THESIS
submitted to
Oregon State University
In partial fulfillment of
the requirement for the
degree of
Master of Science
Presented August 24th, 2004
Commencement June 2005
Master of Science thesis of Methea Sapp presented on August 24th 2004
Redacted for Privacy
Major P/ofçsor, representing Zoology
Redacted for Privacy
Chair o tl1ê Department of Zoology
Redacted for Privacy
Dean of the Grr'duate School
I understand that my thesis will become part of the permanent collection of Oregon State
University libraries. My signature below authorizes release of my thesis to any reader upon
request.
Redacted for Privacy
/ Methea Sapp, Author
ACKNOWLEDGEMENTS
I thank my advisor, John Ruben for challenging me to pursue this body of work. I
also wish to thank the members of my committee, Art Boucot, George Poinar, and
Mark Harmon for their time and expertise. I am also deeply indebted to both Amy
Harwell and Devon Quick for the collegiate spirit with which they so freely gave me
both their time and resources. Jaap Hillenius and Michael Cassanego also freely gave
their time and expertise with numerous edits and suggestions for improvement. Frank
Nelson also has my thanks for his continuous encouragement during the writing
process. I am also thankful to Bob Mason for his assistance and support during the
long process of securing salvage permits. Many other individuals contributed both
their time, expertise and resources in an effort to make this thesis a success, they are;
Jim Glick of Backacres Farm, and Frittz Gottfried of Hawthorn Emu Ranch, both of
whom donated Emu carcasses, Chintimini Wildlife Rescue for the donation of
numerous avian wildlife specimens, and Oregon State University Poultry Barns for
their donation of poultry carcasses. Finally I would like to extend my gratitude to
Brett Gartrell of Massey University for the kiwi CT scans and to Judy and Joe
Passentino for the tinamou specimen.
TABLE OF CONTENTS
INTRODUCTION
MATERIALS AND METHODS
.
................................................
6
RESULTS..........................................................................
9
DISSCUSION
.....................................................................
11
BIBLIOGRAPHY.................................................................
15
LIST OF FIGURES
Figure
1. The avian respiratory system........................................................................... 3
2. Avian thoracic skeletal changes during respiration ..................................................4
3. Abdominal cavity of an adult Merlin .................................................................. 7
4. Lateral view of the thoracic skeleton and pelvis of a Rock Dove (Columbia livia) .............. 8
5. Ventilatory movement of thoracic and abdominal skeletal elements and muscle groups ...... 12
6. Pubic bone orientation in theropods as exemplified by the dromaeosaurid theropod ........... 13
Deinonychus.
7. Pubic bones from; A. Albertasaurus (photograph courtesy of Jaap Hillenius); ................. 14
B. Mircovenator; C. Velociraptor;
DEDICATION
To Mom, Dad, Lysandra, and John C., who sustain me with their everlasting love and support,
and to the late Dr. Howard Towner who ignited my passion for birds.
DINOSAUR LUNG STRUCTURE AND VENTILATION OF THE
ABDOMINAL AIR SACS IN BIRDS
INTRODUCTION
Birds are supposedly derived directly from the theropod dinosaurs. This current dogma
originated in the mid-1970's with Ostrom's descriptions of the similarities between
Deinonychus and Archaeopteryx (Ostrom 1976) and has since progressed to volumes of
works not only supporting the theropod-bird lineage, (Cume 1985, Novas & Puerta 1998,
Forster et al. 1998, Xu et a]. 1999, Gauthier 1986, Norell et al. 2001, Prum 2002, Padian
and Chiappe 1998), but in fact claiming that birds are dinosaurs (Dingus and Rowe 1998;
Paul 1988, 2002).
There are three principle adaptations which define the general anatomical plan of
modern birds. These include skeletal rigidity, skeletal reduction, and redistribution of mass
(Proctor and Lynch 1993). Other attributes of modem Ayes include, but are not limited to,
the presence of a furcula, feathers, long forearms in which the ulna and third metacarpal are
bowed, fused carpometacarpus, laterally flexing wrists, pneumaticized bones, retroverted
pubis, a reversed first toe called the hallux, and the synsacrum. Theropods also possessed
some of these attributes. For example, the small dromaeosaurid, Deinonychus, possessed
elongate forearms, bowed ulna and metacarpal, and the semilunate carpal, all of which are
essential components of the avian wing (Dingus and Rowe 1998). Furthermore,
comparative analyses of Velociraptor, Compsognathus, Coelurus, Microvenator, and
Struthiomimus with Archaeopteryx have revealed a considerable number of supposed
synapomorphies between the theropods and primitive bird (Ostrom 1976).
Proponents of the theropod ancestry of birds have recognized that osteological features
alone do not define a bird. Thus, based on a supposedly close theropod-bird relationship,
many also assume that theropods possessed a "high performance," avian-like lung-air sac
2
respiratory system and, therefore, might have been endothermic (Paul 1988; 2000,
Claussens 1998, Dingus and Rowe 1998). However, phylogenetically-based interpretation
of paleoanatomy has often proven to be broadly misguided and frequently incorrect
(Ruben, et al., 2003). Clearly, a priori, cladistically-based assumptions regarding an avianlike structure for theropod lungs should be regarded cautiously. Rather, a more powerful
argument for bird-like lungs in dinosaurs must necessarily be based on theropods having
possessed specialized skeletal attributes which are also present in living birds and which are
essential to the function of the avian air sac lung. Alternately, the absence of such skeletal
specializations in dinosaurs might serve to falsify assumptions that theropod dinosaurs
possessed avian-style lungs.
Birds are unique among all extant air-breathing vertebrates in that the avian lung is
ventilated both continuously and unidirectionally, so that air flows ad infinitum through the
highly compartmentalized parenchyma and air sacs (Figure 1). The vascularized region of
the lung (= parenchyma), in combination with a total of nine air sacs (interclavicular and
the paired cervical, anterior thoracic, posterior thoracic, and abdominal) occupies 20% of
the total body volume in birds. This volume is three to five times greater than in mammals
and twice that in reptiles (Mama, 2002). The unidirectional air flow pattern of the avian
lung allows for varying P02 and PCO2 within the air sacs. Since they receive gas from the
parabronchi, the cranial air sacs, (cervical, and interclavicular) contain relatively high
PCO2 levels, and low P02. In contrast, the caudal air sacs (posterior thoracic, and
abdominal sacs) receive a combination of re-inhaled and fresh air which elevates their P02
and lowers their PCO2. Upon exhalation it is this 02 rich volume of air that exits the caudal
air sacs to ventilate the parabronchi (Powell, 2000). It has long been recognized that
the flow-through avian lung cannot fully function in the absence of well-developed
abdominal air sacs (Brackenbury 1990).
3
Figure 1: The avian respiratory system. Arrows indicate the pathway of a single breath of
air. The first inhalation (posteriorly directed black arrows), pulls air past the lung and into
the abdominal air sacs. Upon the first exhalation (anteriorly directed black arrow), the air
is pushed from the abdominal air sacs into the parabronci. Upon the second inhalation
(anteriorly directed white arrow), air is pulled from the lungs and into the anterior air sacs.
Finally, the second exhalation (last two white arrows), pushes air out of the anterior air sacs
and into the trachea. It is important to note that during the two-part respiratory cycle, the
thoracic volume is increasing and decreasing as a result of skeletal movements (adapted
from Proctor and Lynch 1993).
Ventilation of the avian lung/air sac system results from a series of changes in the
functional position
of
the thoracic skeleton during inspiration and expiration (Figure 2).
During inspiration the sternum is rotated both cranially and ventrally while the thoracoabdominal cavity and the sternal ribs are expanded laterally via cranial movement of the
vertebral ribs. This skeletal shift increases the volume of the thoracic and abdominal
cavities and creates negative internal pressure which results in filling of the respiratory
system.
However, under conditions of negative thoracic and abdominal pressures, any
unsupported, soft internal structures not held in place or maintained in an open
fl
Ii
_'\
Ik
\
El
1k
Ik
n i.-4L-----1Ir7/
fZ
11'
Hit
I/li
IIti
Flu
I1
ffi'
II%
U'
U t 1111
II%% lit'
Ii
II
Fl
LI \
ll
II'
II \
'\
\\
U" U\\ U \It \\U
U \'
U
U \ It
U " U \1
'1I'
I!'
II
!f
1/,,
liii
1/ /
A21
'
\\1__......2
I
1
\\
\\
/
'-a.--..
..x4*..çe
I
/
/
Sternum
Figure 2: Avian thoracic skeletal changes during respiration. Dashed lines indicate skeletal
position during inhalation; while the solid lines depict skeletal orientation during
exhalation. These skeletal shifts change both thoracic and abdominal pressures and are
essential to avian respiration (adapted from Kardong 2002).
position, (e.g., the air sacs) are subject to shift and/or collapse. This phenomenon, termed
"paradoxical movement," is of particular significance, especially as it pertains to the
voluminous abdominal air sacs. These are paired, dorsally situated "air bladders" which are
located in the abdominal cavity, largely posterior to the costal skeleton, which is well suited
only to prevent paradoxical movements of the clavicular and thoracic air sacs. Curiously,
even though abdominal air sacs are essential components of the avian lung, there are no
previously described osteological features linked to their skeletal support during avian lung
ventilation.
I investigated whether there is an unrecognized combination of skeletal attributes that
prevent paradoxical inhalatory movements of the abdominal air sacs in birds. More
specifically, I hypothesize that during inhalation, specialized pelvic skeletal elements,
5
especially the synsacrum and the fully retroverted and unfused pubic bones, act in concert
to stabilize these air sacs and prevent their collapse. Abdominal air sacs are located in
close proximity to these axial skeletal structures and might reasonably be expected to
derive ventilatory support from them. Furthermore, if my hypothesis is correct, the
presence or absence of similar skeletal attributes in theropod dinosaurs might provide
evidence for, or against, the former existence of abdominal air sacs in theropods and, by
extension, the presence of an avian-style lung air sac system in those animals.
MATERIALS AND METHODS
I investigated (via simple dissection) the structure and placement of abdominal air sacs
and their closely associated axial skeletal elements (i.e., synsacrum, and pubic bones) in a
combination of 26 fresh and frozen adult bird carcasses from 10 distinct avian orders.
These included: Anseriformes: Northern Shoveler, Green-winged Teal; Passeriformes:
Stellars Jay, Barn Swallow, American Robin; Columbiformes: Rock Dove; Falconiformes:
Sharp-shinned Hawk, Merlin, Red-tailed Hawk, American Kestral; Strigiformes: Western
Screech Owl, Barn Owl, Great Horned Owl; Tinamiformes: Chilean Tinamou;
Struthioniformes: Emu; Picciformes: Sapsucker; Pelicaniformes: Double-crested
Cormorant, Brown Pelican; Galliformes: domestic Chicken.
All non-domesticated specimens were donated from Chintimini Wildlife Rescue Center
and registered under Federal and State Salvage permits. Poultry were donated by Oregon
State University's Department of Animal Sciences; Emu specimens came from Backacres
Emu Ranch of Olympia, Washington and Hawthorn Emu Ranch in Tacoma, Washington.
To insure anatomical accuracy, no bird which had suffered thoracic or abdominal
trauma was included in this study. All dissections were performed from the ventral side
with
birds prone (Figure 3). Abdominal air sacs in dead specimens are inevitably collapsed
with little indication of their true anatomy and function. Therefore, for these studies it was
necessary to artificially inflate the respiratory system to accurately define its true
anatomical relationship to the axial skeleton. To facilitate inflation of the respiratory
system, an incision was made in the lower trachea. A section of plastic tubing was then
fitted to closely match the diameter of each specimen's trachea; this was then inserted into
the tracheal slit. The respiratory system was then fully inflated by gently blowing into the
tube; this was subsequently sealed using a pair of forceps. To reduce the chance of
Figure 3: Abdominal cavity of an adult Merlin. Note the extensive occupation by the
thoracic and abdominal air sacs (th-as and ab-as, respectively). Since the caudal air sacs
exist in pairs, arrows point to the corresponding air sac situated on the opposite side of the
body. A thin sheet of abdominal muscle has been left partially intact on the animals left
side (ab-ms). The medial tip of both the left and right pubic bones are indicated by (pb).
distending or overfilling of the air sacs beyond expected "in situ" capacity, the abdomen
was kept intact prior to dissection of the abdominal air sacs.
Once the respiratory system was inflated the superficial abdominal tissues were
carefully removed to the extent that both the lateral and caudal edges of the abdominal air
sacs and pubic bones were observed. The abdominal air sacs were observed to invariably
extend posteriorly from the ribcage into the abdominal region and to be located dorsally
within the synsacrum and laterally to be bounded by the pubic bones. Removal of the
infrapubic and suprapubic abdominal musculature revealed extensive connective tissue
which tightly fused the air sacs to the aforementioned musculature as well as to the medial
aspects of the pubic bones, synsacrurn and, in some cases, the gizzard. A series of two
measurements were then recorded for each specimen (Figure 4). Distances were recorded
in millimeters using a pair of Whitworth electronic digital calipers. These measurements
included: The length from the posterior tip of the sternum to the end of the abdominal airsacs st-as; and the length from the posterior tip of the sternum to the tip of the pubis st-pb.
The proportion of longitudinal abdominal air sac length supported by the pubic bones
and synsacrum (=[st-pb/st-as]) was then calculated.
Figure 4: Lateral aspect of the thoracic skeleton and pelvis of a Rock Dove (Columba
livia). Illustrates skeletal structures and distances measured. Sternum (St); Sternal rib (Sr);
Vertebral rib (Vr); Abdominal air sacs (As); Ischium (Is); ilium (Ii); Pubis (Pb); The
length from the posterior tip of the sternum to the end of the abdominal air-sacs (st-as);
Length from the posterior tip of the sternum to the tip of the pubis (st-pb). (adapted from
Proctor and Lynch 1993).
Data was compiled using a simple Microsoft Excel spread sheet so that ratios of the
distance between structural markers were then calculated.
RESULTS
Primary findings from dissection measurements indicate that, in every specimen in this
study, the lateral post-sternal extension of the synsacrum+pubic bones coincides closely
with the post-sternal extension of the abdominal air sacs. In fact, the percent of dorso-
lateral "coverage" of the abdominal air sacs by the combination of synsacrum and pubic
bones was 79% or greater in every case (Table I).
Visual inspection and dissection revealed that the dorsal surfaces of the abdominal air
sacs are tightly fused to the ventral surface of the synsacrum and the medial aspects of the
pubic bones via the parietal peritoneum. Ventrally, the abdominal air sacs adhere tightly
(via connective tissue) to both the suprapubic and infrapubic abdominal musculature.
Additionally, as is true for Ayes in general, pubic bones in every specimen in this study
were widely separated and lack a symphysis. Therefore, the unfused and broadly spaced
pubic bones allowed for the requisite space for the voluminous abdominal air sacs.
10
Table 1: Attributes of skeletal and soft tissues associated with the abdominal air-sacs in 10
avian orders. Dissection measurements in millimeters, include the lengths between the
sternal tip and the posterior extremity of the abdominal air sacs (st-as), and the length
between the sternal tip and tip of the pubic bone (st-pb); The percentage of abdominal air
sac support dorsally and laterally by the pubic bones and synsacrum is expressed as the
ratio of st-pb to st-as (see third data column). These data reveal a strong correlation
between the length of the pubis and the extent of the abdominal air sacs.
Percent of abd. air sac length
supported dorsally and laterally by
the synsacral/pubic complex
(=st-pb/st-as)
st-as
st-pb
(mm)
(mm)
67.79
67.30
23.86
55.42
60.32
18.92
82
90
79
30.73
10.83
19.17
23.91
29.14
9.26
16.28
20.24
20.86
19.61
95
85
85
85
94
18.06
16.46
91
23 00
34.68
52 06
29.61
22 88
36.75
48 29
30.43
100
93
100
23.86
22 48
52.56
64 89
27.04
26 29
51.86
58 43
20.67
13.18
16.46
14.25
Emu
378.00
387.00
100
Piccuformes
Sapsucker
Pelican iformes
13.40
12.26
91
120.27
107 22
125.36
93 05
100
33.10
31.73
24.18
27.82
96
100
Bird
Anseriformes
Northern Shoveler 1
Northern Shoveler 2
Green-winged Teal
Passeriformes
Stellars Jay
Barn Swallow
American Robin 1
American Robin 2
American Robin 3
Columbiformes
Rock Dove
Falcon iformes
Sharp Shinned Hawk
Merlin
Red tailed Hawk
American Kestral
99
Stringiformes
Western Screech Owl 1
Western Screech Owl 2
Barn Owl
Great Horned Owl
100
100
99
90
Tinamiformes
Tinamou 1
Tinamou 2
.
80
100
Struthioniformes
Pelican
Cormorant (dbl crested)
Gall iform es
Chicken 1
Chicken 2
Chicken 3
23.81
20.17
87
100
11
DISCUSSION
For a large fraction of their antero-posterior length, the abdominal air-sacs are fused
tightly to, and supported dorsally and laterally by the synsacrum and pubic bones in every
bird included in this study (Table 1: st-as/st-pb). Of the 26 specimens examined here, all
possessed pubic bones which "covered" at least 79% of the length of the abdominal air
sacs. Thus, there exists a strong relation between the caudal extent of the abdominal air
sacs and the degree of dorsal and lateral support provided by the synsacrum and pubic
bones.
th addition to the lateral support of the abdominal air sacs provided by the pubic bones,
paradoxical inhalatory movement of these air sacs is undoubtedly prevented by the
infrapubic and suprapubic musculature ventrally (Figure 5). The abdominal air-sacs adhere
to the ventral and lateral sides of the synsacrum via fusion of the parietal peritoneum of the
body wall to the peritoneal coverings of the air-sacs, thereby securing them both dorsally
and dorso-laterally (McLelland 1989). Ventral support is provided by infrapubic
musculature, which is fused to the peritoneal coverings of the air-sacs.
Although air sac support by pubic musculature has not been noted previously, Baumel
reports that contraction of this musculature coincides with inhalation. Consequently, fusion
between the fascia of the suprapubic abdominal muscles and ventral portions of the
abdominal air-sacs probably prevent the ventral paradoxical movement of these air sacs
during inhalation.
These parameters provide evidence that in extant birds, resistance of the abdominal air
sacs to lateral, dorsal, and ventral inhalatory paradoxical movement is provided by the
pubic bones, synsacrum, and abdominal musculature. Extinct taxa with avian style, air sac
lungs would therefore, be expected to have exhibited similar osteological attributes to those
described here.
12
M. on iimu
dQI
/
\
oarium
/
Syrsc-nni M cudofemorfj
Postnacta
\
interveflcbr& joint
------\ -------
\\\
/
1im
(7
,.' \
/--
(\tV/_
I
II/J//J!i?/
t
N
Sum
/
-
-
I
I
Suprapubc
biominaJ
mudes
Infrabk
bdomini
Figure 5. Ventilatory movements of thoracic and abdominal skeletal elements and muscle
groups. In addition to aiding in respiration, the infrapubic abdominal muscles and the
suprapubic abdominal muscles prevent the ventral and caudal paradoxical movement of the
abdominal air sacs via fusion of the peritoneal linings. The air sacs are also supported
dorsally by the synsacrum and laterally by the pubic bones. Dashed lines and arrows
indicate movement of sternum and pelvis during inspiration. Note that these movements do
not occur simultaneously (From Baumel, et.al. 1990).
In sharp contrast to previous assumptions of avian style respiration in some dinosaurs,
these data suggest that theropods probably did not possess the essential osteological
characteristics necessary to ventilate abdominal air sacs and by extension to support a high
performance avian lung air sac system. As described above there are a number of
osteological attributes that seem required for the support and ventilation of abdominal air
sacs: 1. A fully retroverted pubis whose long axis is orientated essentially parallel to the
lateral aspect of the abdominal air sacs; 2. the presence of a synsacrum, and broadly open
pubic bones through which abdominal air sacs may extend.
13
With few exceptions, theropod pubic bones are orientated vertically, relative to the hip
joint (Figure 6). Therefore, the right-angle positioning of the pubic bone could not have
provided substantial lateral support for avian style abdominal air sacs. Furthermore, within
these extinct taxa, the complete absence of a synsacrum is also inconsistent with an ability
to have prevented dorsal and dorso-lateral paradoxical movements of abdominal air-sacs
Figure 6. Pubic bone orientation in theropods as exemplified by the dromaeosaurid
theropod Deinonychus. (Illustration adapted from Carroll 1988) Note the vertical
orientation of the pubic bones (Pubic bones are shaded gray).
Finally, no known theropod possessed the unfused, broadly spaced pubic bones typical
of modern birds. in contrast, the pubic bones of theropods were characterized by a lengthy
medial fusion of the two pubic halves which resulted in formation of a broad, uninterrupted
osseous pubic apron (Figure 7). This elongate pubic symphysis is inconsistent with the
presence of an avian style abdominal air sac, which in modern birds inevitably passes
between and extends posteriorly through the pubic rami. Thus, even if orientation of the
pubic bones were fully retroverted there would likely have been inadequate space between
the pubic rami to contain the colon, cloacal structures, and abdominal air-sacs. Based upon
these data, it is clear that proper ventilation of abdominal air sacs in the avian lung requires
14
Iv'
A.
B.
C.
Figure 7. Pubic bones from: A. Albertasaurus (photograph courtesy of Jaap Hillenius);
B. Mircovenator; and C. Velociraptor; Note that the fused pubic apron in specimens A-C
do not provide space for abdominal air-sacs.
a distinct combination of anatomical features which seem present in all extant birds, but
which were apparently absent in all theropods. These data do not support previous
assumptions that theropods ventilated an avian-lung air sac system.
15
BIBLIOGRAPHY
Baumel, J. J., Wilson, J. A., and Bcrgren, D. R. The ventilatory movements of the
avian pelvis and tail: Function of the muscles of the tail region of the pigeon
(Columba livia). J. Exp. Biol. 151: 263-277 (1990).
Brackenbury, J. H., and Amaku, J. Effects of Combined Abdominal and Thoracic
Air-Sac Occlusion on Respiration in Domestic Fowl. J. exp.Biol. 152: 93-100
(1990).
Brackenbury, J. H. Ventilation of the lung-air sac system. In Seller, T. J. Bird
Respiration, vol. 1: 39-69 (CRC Press, Inc.: Boca Raton, 1987).
Carroll, L. R. Vertebrate Paleontology and Evolution. (W. H. Freeman and
Company: New York, 1988).
Claussens, L. P. A. M., Perry, S. F., and Currie, P. J. Using comparative anatomy to
reconstruct theropod respiration. J. Vert. Paleontol. 18 (suppi. to 3): 34A
(1998).
Dingus, L., and Rowe, T. The mistaken extinction: Dinosaur evolution and origin
of birds. (W. H. Freeman and Company: New York, 1998).
Forster, C. A., Sampson, S. D., Chiappe, L. M., and Krause, D. W. The theropod
ancestry of birds: new evidence from the Late Cretaceous of Madagascar.
Science 279: 1915-1918 (1998).
Gauthier, J. A. Saurischian monophyly and the origin of birds. In Memoir of the
California Academy of Sciences. Edited by K. Padin. 8: 1-55 (1986).
Kardong, K. V. Vertebrates: comparative anatomy, function, evolution. (McGrawHill Companies Inc.: New York, 2002).
Mama, J. N. Functional Morphology of the Vertebrate Respiratory Systems.
(Science Publishers Inc.: Enfield, 2002).
McLelland, J. Anatomy of the lungs and air sacs. In Form and Function in Birds.
Edited by A. S. King and J. McLelland (Academic Press: New York, 1989).
Norell, M., Clark, J. M., and Makovicky, P. J. Phylogenetic relationships among
coelurisaurian dinosaurs. In New perspectives on the origin and early evolution
of birds, edited by J. Gauthier, and J. F. Gall. (Peabody Museum of Natural
History: New Haven, 2001).
16
Ostrom, J. 1-1. Archaeopteryx and the origin of birds. Biolo. J. Linn. Soc. (Lond.) 8:
91-182 (1976).
Padian, K. and Chiappe, L. M. The origin and early evolution of birds. Biol. Rev.
73: 1-42 (1998).
Paul, G. S. Predatory dinosaurs of the world. (Simon and Schuster: New York,
1988).
Paul, G. S. Dinosaurs of the Air. (John Hopkins University Press: Baltimore,
2002).
Powell, F. L. Respiration. In Sturkie's Avian Physiology. Edited by G. Causey
Whittow (Academic Press: New York, 2000).
Proctor N. S. and Lynch. P. J. Manual of Ornithology: Avian Structure and
Function. (Yale University Press: New Haven, 1993).
Prum, R. 0. Why ornithologists should care about the theropod origin of birds. Auk
119: 1-17(2002).
Ruben, J. A., Jones, T. D., and Geist, N. R. Respiratory and Reproductive
Paleophysiology of Dinosaurs and Early Birds. Physiological and Biochemical
Zoology 76: 141-164(2003).
Xing, X. U. Four-winged dinosaurs from China. Nature 421: 335-340 (2003).